Low Urine pH: A Novel Feature of the Metabolic Syndrome

Study Participants

We used local advertisements to recruit volunteers who were willing to participate in the study. Inclusion criteria were age >21 yr and ability to provide informed consent. We wished to evaluate individuals with a wide range of body weight; therefore, body weight and/or body mass index (BMI) was not used for the inclusion/exclusion criteria. Excluded from the study were individuals with conditions that are known to alter urine pH, such as CKD (defined by creatinine clearance <70 ml/min [<1.17 ml/s]), renal tubular acidosis, chronic diarrheal illness, kidney stone disease (by history), primary hyperparathyroidism, or urinary tract infections. Participants were instructed to hold any medications that are known to influence urine pH (e.g., alkali therapy, diuretics) for 1 wk before the study. Individuals with type 2 diabetes were allowed to participate if they were not taking insulin or thiazolidinediones. The study was approved by the institutional review board at the University of Texas Southwestern Medical Center (Dallas, TX).

Data Collection and Measurements

All participants collected a 24-h urine sample under mineral oil, and the urine container was kept refrigerated or maintained in an ice chest during collection. At the end of the urine collection, the participants had fasting blood drawn, and height, weight, and BP were measured. The American Heart Association/National Heart, Lung, and Blood Institute updated the definition of the metabolic syndrome in 2005 (15) to the presence of three or more of the following: (1) An elevated fasting blood glucose (≥100 mg/dl, ≥6 mmol/L) or drug treatment for elevated glucose; (2) an elevated BP defined as systolic BP (SBP) >135 mmHg, diastolic BP >85 mmHg, or drug treatment for hypertension; (3) elevated fasting serum triglycerides defined as >150 mg/dl (>1.70 mmol/L) or drug treatment for hypertriglyceridemia; (4) low fasting serum HDL cholesterol <50 mg/dl (<1.30 mmol/L) in women and <40 mg/dl (<1.04 mmol/L) in men or drug treatment for low HDL; and (5) abdominal obesity defined as waist circumference >88 cm in women and >102 cm in men. Because waist circumference was not available for many individuals, a BMI of ≥30 kg/m² was used as an index of obesity. BMI is considered a satisfactory substitute for waist circumference because the two parameters are highly correlated (16). BMI has been used instead of waist circumference in modified criteria for the definition of the MS in several previous studies (17,18). Insulin resistance was calculated from fasting glucose and insulin measurements using the homeostasis model assessment for insulin resistance (HOMA-IR) (19,20).

Analytical Procedures

Serum electrolytes, glucose, triglycerides, total and HDL cholesterol, creatinine, and uric acid concentrations were measured using an automated system (Beckman CX9ALX, Fullerton, CA). Urine pH was measured by pH electrode. Urine creatinine was determined using the picric acid method, and endogenous creatinine clearance was calculated from values of creatinine in the serum and urine. Urine sulfate was evaluated by ion chromatography. Urine bicarbonate (HCO₃⁻) was calculated from urine pH and Pco₂. Urine ammonium (NH₄⁺) was determined by the glutamate dehydrogenase method, and urine titratable acidity was measured directly using the automated burette end point titration system (Radiometer, Copenhagen, Denmark). Urine citrate was evaluated enzymatically using reagents from Boehringer-Mannheim Biochemicals (Indianapolis, IN), and milliequivalents of citrate were calculated from urine pH and a pKa of citrate²⁻/citrate³⁻ of 5.6. Urine net acid excretion (NAE) was calculated as urine (NH₄⁺ + titratable acidity) − (citrate + HCO₃⁻), all in milliequivalents. The proportion of net acid excreted as ammonium is described by the ratio of NH₄⁺ to NAE (NH₄⁺/NAE). Serum insulin was assessed by ELISA (Mercodia, Metuchen, NJ).

Statistical Analyses

Categorical comparisons were performed using the χ² test. Serum triglycerides and HOMA-IR were log-transformed before statistical analyses. For normally distributed continuous variables, comparisons of participants with and without MS were made with two-sample t test. For urine NH₄⁺/NAE, the Wilcoxon rank sum test was used. Tests of linear trend were conducted by one-way ANOVA with polynomial contrasts. The urine NH₄⁺/NAE trend was analyzed with the Jonckheere-Terpstra test. Analysis of covariance models were used to examine the relationship between urinary pH with the number of MS abnormalities and for assessing each MS feature while adjusting for the potential confounding effects of covariates. BMI, age, gender, urine sulfate, and creatinine clearance were included as covariates. Statistical analyses were performed using SAS 9.1.3 (SAS Institute, Cary, NC).

Demographic Characteristics

A total of 148 individuals participated in the study. The demographic characteristics of the entire cohort are shown in Table 1. The average age (±SD) was 44 ± 14 yr, and the mean BMI was 28.4 ± 6.7 kg/m². The average number of MS features was 1.76 ± 1.51, and 29.7% of the cohort met the definition of the MS.

Urine pH and the MS

The demographic and metabolic characteristics of participants with (MS+) and without (MS−) the MS are shown in Table 2. There was no significant difference in the racial distribution between MS+ and MS− participants. MS+ participants were significantly older and had significantly greater BMI, SBP and diastolic BP, and serum glucose and triglycerides. Participants with the MS had a significantly lower 24-h urine pH compared with participants without the MS (5.71 ± 0.47 versus 6.11 ± 0.42; P < 0.001). This significant difference in urine pH between MS+ and MS− participants was noted in both genders. The 24-h urine potassium was not different between the two groups. MS+ participants had significantly greater NH₄⁺ and NAE, although NH₄⁺/NAE was significantly lower than in MS− participants.

Urine pH and Features of the MS

Participants were further classified according to the number of metabolic abnormalities (Table 3). Mean urine pH according to the number of characteristics of the MS is shown in Table 3. Mean 24-h urine pH values decreased with increasing number of MS features (from 6.19, 6.14, 5.96, 5.83 to 5.60; P < 0.0001 for trend). Age, creatinine clearance, and 24-h urine sulfate (a surrogate for dietary acid ingestion) were also significantly correlated with urine pH in univariate analysis (P < 0.05). These variables were included in a multivariate model to examine the relationship between urine pH and the MS. Mean 24-h urine pH adjusted for age, gender, creatinine clearance, and urine sulfate decreased from 6.15, 6.10, 5.99, 5.85, to 5.69 with increasing number of metabolic abnormalities (P < 0.005 for trend; Figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Mean 24-h urine pH adjusted for age, gender, creatinine clearance, and 24-h urine sulfate plotted against number of features of the metabolic syndrome (MS). Bars indicate means ± SEM. P < 0.005 for linear trend. Test of linear trend was conducted by analysis of covariance model comparing urine pH with the number of MS abnormalities while adjusting for age, gender, creatinine clearance, and 24-h urine sulfate.

Twenty-four-hour urine pH was significantly correlated with serum HDL cholesterol and inversely and significantly associated with BMI, SBP, serum glucose, and serum triglycerides (Table 4). After adjustment for age, gender, creatinine clearance, and urine sulfate, 24-h urine pH was still significantly associated with BMI, SBP, serum glucose, and serum HDL cholesterol, although the magnitude of the association decreased (Table 4).

Urine pH and Insulin Resistance

Twenty-four-hour urine pH was inversely correlated with the degree of insulin resistance estimated by the HOMA-IR (P < 0.0001; Table 4). After adjustment for age, gender, BMI, creatinine clearance, and urine sulfate, the inverse relationship between 24-h urine pH and HOMA-IR remained statistically significant (Table 4).